{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CMBMQHDP2J227NCQ27CJFN7M5P","short_pith_number":"pith:CMBMQHDP","schema_version":"1.0","canonical_sha256":"1302c81c6fd275afb450d7c492b7ecebecc3d7eec1f3739107790544bbc67854","source":{"kind":"arxiv","id":"2509.08266","version":1},"attestation_state":"computed","paper":{"title":"Examining Vision Language Models through Multi-dimensional Experiments with Vision and Text Features","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Donald E. Brown, Jiebei Liu, Nazanin Moradinasab, Saurav Sengupta","submitted_at":"2025-09-10T03:49:40Z","abstract_excerpt":"Recent research on Vision Language Models (VLMs) suggests that they rely on inherent biases learned during training to respond to questions about visual properties of an image. These biases are exacerbated when VLMs are asked highly specific questions that require focusing on specific areas of the image. For example, a VLM tasked with counting stars on a modified American flag (e.g., with more than 50 stars) will often disregard the visual evidence and fail to answer accurately. We build upon this research and develop a multi-dimensional examination framework to systematically determine which "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2509.08266","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-09-10T03:49:40Z","cross_cats_sorted":[],"title_canon_sha256":"cc6c826bdaa00ec1b4d37818dd5718cbc80efcb5577b780f04409f6e0d4feaa5","abstract_canon_sha256":"d5ad43750473e713678711c023b830f21499a1f717fb063ec73aa41406eedecf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:08:26.549805Z","signature_b64":"zuxzdLZcsaK27QlZQd+xN6SM0aDhdneuOs2amynvlNzzyPwQdC6pMwKfsCacx58MvgZSVpmxUMCc+socSe83AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1302c81c6fd275afb450d7c492b7ecebecc3d7eec1f3739107790544bbc67854","last_reissued_at":"2026-07-05T12:08:26.549271Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:08:26.549271Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Examining Vision Language Models through Multi-dimensional Experiments with Vision and Text Features","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Donald E. Brown, Jiebei Liu, Nazanin Moradinasab, Saurav Sengupta","submitted_at":"2025-09-10T03:49:40Z","abstract_excerpt":"Recent research on Vision Language Models (VLMs) suggests that they rely on inherent biases learned during training to respond to questions about visual properties of an image. These biases are exacerbated when VLMs are asked highly specific questions that require focusing on specific areas of the image. For example, a VLM tasked with counting stars on a modified American flag (e.g., with more than 50 stars) will often disregard the visual evidence and fail to answer accurately. We build upon this research and develop a multi-dimensional examination framework to systematically determine which "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.08266","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2509.08266/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2509.08266","created_at":"2026-07-05T12:08:26.549349+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.08266v1","created_at":"2026-07-05T12:08:26.549349+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.08266","created_at":"2026-07-05T12:08:26.549349+00:00"},{"alias_kind":"pith_short_12","alias_value":"CMBMQHDP2J22","created_at":"2026-07-05T12:08:26.549349+00:00"},{"alias_kind":"pith_short_16","alias_value":"CMBMQHDP2J227NCQ","created_at":"2026-07-05T12:08:26.549349+00:00"},{"alias_kind":"pith_short_8","alias_value":"CMBMQHDP","created_at":"2026-07-05T12:08:26.549349+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CMBMQHDP2J227NCQ27CJFN7M5P","json":"https://pith.science/pith/CMBMQHDP2J227NCQ27CJFN7M5P.json","graph_json":"https://pith.science/api/pith-number/CMBMQHDP2J227NCQ27CJFN7M5P/graph.json","events_json":"https://pith.science/api/pith-number/CMBMQHDP2J227NCQ27CJFN7M5P/events.json","paper":"https://pith.science/paper/CMBMQHDP"},"agent_actions":{"view_html":"https://pith.science/pith/CMBMQHDP2J227NCQ27CJFN7M5P","download_json":"https://pith.science/pith/CMBMQHDP2J227NCQ27CJFN7M5P.json","view_paper":"https://pith.science/paper/CMBMQHDP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.08266&json=true","fetch_graph":"https://pith.science/api/pith-number/CMBMQHDP2J227NCQ27CJFN7M5P/graph.json","fetch_events":"https://pith.science/api/pith-number/CMBMQHDP2J227NCQ27CJFN7M5P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CMBMQHDP2J227NCQ27CJFN7M5P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CMBMQHDP2J227NCQ27CJFN7M5P/action/storage_attestation","attest_author":"https://pith.science/pith/CMBMQHDP2J227NCQ27CJFN7M5P/action/author_attestation","sign_citation":"https://pith.science/pith/CMBMQHDP2J227NCQ27CJFN7M5P/action/citation_signature","submit_replication":"https://pith.science/pith/CMBMQHDP2J227NCQ27CJFN7M5P/action/replication_record"}},"created_at":"2026-07-05T12:08:26.549349+00:00","updated_at":"2026-07-05T12:08:26.549349+00:00"}